What Google's new GEO guide actually means for healthcare practices.

Google published their official guide to optimizing for generative AI in Search on May 15th, and my feed immediately filled up with takes ranging from "everything changes now" to "nothing to see here, keep doing SEO." Neither of those is quite right, and I think the nuance matters a lot if you're a healthcare practice trying to figure out where to spend your time.

Here's my read on it, specifically for specialty medical practices.

First, a distinction most coverage is skipping

The Google document is about Google's AI features: AI Overviews and AI Mode. It is not about ChatGPT, Perplexity, Claude, or Gemini. Those are different systems with different architectures, different citation behaviors, and different content signals. What works to get cited in Perplexity is not identical to what Google is describing in this guide.

I say this because I've seen a lot of "GEO is dead, Google says just do SEO" reactions that conflate all AI search into one thing. There's real intentional work required to show up across all of these systems — and each one has different signals worth understanding.

That said, Google's guide covers their territory, and Google is still where the majority of healthcare patient searches happen. So it deserves a close read.

What Google actually said

The core message is this: appearing in AI Overviews doesn't require special optimization beyond good, legitimate SEO. No llms.txt files. No "chunking" your content into AI-digestible pieces. No special schema markup. No manufactured mentions across the web to inflate your citation footprint.

What they do want is non-commodity content. They're explicit about this. They draw a line between content that rehashes common knowledge (the kind of page a hundred practices could publish with minimal effort) and content that reflects actual experience and specific insight. The latter gets cited. The former doesn't stand out enough to.

For healthcare practices, this is actually good news. Your clinical experience is the differentiator, if you put it on the page. A rhinoplasty surgeon who writes specifically about how they approach the deviated septum that also affects external appearance, with case-level details and realistic outcome framing, has something that no content farm can replicate. That's what Google is describing as citation-worthy.

The piece I'm watching most closely

Google introduced a section on "agentic experiences" — AI agents that browse and interact with websites on behalf of users. Book appointments. Compare options. Fill out forms. Complete tasks that patients currently do manually.

They link to a guide on building agent-friendly websites and mention the Universal Commerce Protocol as an emerging standard for AI-agent interactions.

In healthcare, this is real. "Book me a consultation with a rhinoplasty surgeon in Chicago who takes Blue Cross" is the kind of query AI agents are going to complete — not just search for. Practices whose websites make it easy for an agent to find, evaluate, and take action are going to have a structural advantage over practices that don't.

Most healthcare websites are not built for this right now. Worth keeping an eye on.

What this means for your practice

If you're a specialty practice wondering what to actually do with all of this:

The work doesn't change as much as the framing suggests. Publish specific, experience-based content about the procedures you do. Make sure your Google Business Profile is complete and your reviews are current. Be findable in the directories where Google pulls authority signals — Healthgrades, the ASPS directory, Doximity, your hospital's provider page.

The intentionality matters. "Just do good SEO" undersells it. Practices that are thinking about which questions AI systems are likely to answer about their specialty — and making sure their content answers those questions specifically and credibly — are going to show up more consistently than practices that publish generic procedure descriptions and assume the algorithm will sort it out.

Google's guide doesn't undercut that work. It describes it.

Sources

1. Google Search Central — Optimizing for Generative AI Features on Google Search — Official Google documentation, May 2026.
2. Google Search Central Blog — A new resource for optimizing for generative AI in Google Search — Published May 15, 2026.
3. web.dev — AI Agent Site UX Best Practices — Google's guidance on building agent-friendly websites.